Beyond Equipment Distribution in Needle and Syringe Programmes

Treloar et al. Harm Reduction Journal (2016) 13:18 DOI 10.1186/s12954-016-0107-0

RESEARCH Open Access Beyond equipment distribution in Needle and Syringe Programmes: an exploratory analysis of blood-borne virus risk and other measures of client need Carla Treloar*, Limin Mao and Hannah Wilson

Abstract Background: Despite high levels of equipment distribution through Needle and Syringe Programmes (NSPs) in Australia, the levels of reuse of equipment among people who inject drugs remain concerning. This paper used an exploratory analysis to examine the needs of NSP client that could be addressed by NSPs to enhance service impact and blood-borne virus risk practices. Methods: People who inject drugs were recruited from six NSP sites in Sydney, Australia, to undertake a self-completed survey. Results: Using the responses of 236 NSP client participants, three factors were identified in an exploratory factor analysis: recent risky injection (Eigenvalue 3.63, 20.2 % of variance); disadvantage and disability (Eigenvalue 2.26, 12. 5 % of variance); and drug use milieu (Eigenvalue 1.50, 8.4 % of variance). To understand the distribution of these factors, the standardised factor scores were dichotomised to explore those participants with ‘above average’ vulnerability on each factor. A small group of NSP clients reported a cluster of vulnerability measures. Most participants (55.5 %) reported vulnerability on none or only one factor, indicating that 45.5 % could be considered as having double (35.6 %) or triple (8.9 %) vulnerability. Conclusions: These results challenge NSPs to understand the heterogeneity among their client group and develop programmes that respond to their clients’ range of needs beyond those immediately associated with blood-borne virus (BBV) risk. This paper contributes to the growing evidence base regarding the need for BBV prevention efforts to examine strategies beyond equipment distribution. Keywords: Needle and Syringe Programme, Blood-borne virus risk, Injecting drugs, Equipment reuse

Background the provision of sterile injecting equipment, opiate substi- A key response to the transmission of blood-borne viruses tution treatment and access to HIV treatments [3]. These (BBVs) among people who inject drugs (PWID) is the interventions, while undoubtedly important, cannot ad- provision of sterile injecting equipment [1, 2]. Providing dress the broader range of variables shown to impact BBV injecting equipment, while necessary, does not adequately risk and cannot account for inter-relationships between address all of the needs of PWID. However, there is little these variables. This leaves NSPs potentially delivering less guidance in the literature on the extent and variety of than optimal services if their service delivery model con- Needle and Syringe Programme (NSP) client needs and the centrates solely on access to sterile injecting equipment association of these with BBV risk practices. In the field of without considering broader client needs, particularly in a HIV and injecting drug use, where the evidence is stron- context such as Australia with high levels of distribution of gest, combined prevention interventions are considered as equipment. Australia has world-leading rates of equipment distri- * Correspondence: [email protected] bution for PWID [4]. However, reuse of injecting Centre for Social Research in Health, UNSW, Sydney 2052, NSW, Australia

© 2016 The Author(s). Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Treloar et al. Harm Reduction Journal (2016) 13:18 Page 2 of 8

equipment remains a critical concern even among PWID sites, which were purposefully selected to represent the attending NSPs. Approximately one in four Australian range of publicly funded NSP services in this area, included NSP clients reused their own equipment in the previous two primary NSPs (stand-alone services with specialist month, about 16 % shared others’ needles and syringes staff) with co-located vendingmachinesites,twoprimary (receptive needle and syringe sharing) and nearly 30 % NSP-only sites, and two secondary NSPs (where equipment shared others’ injecting equipment (receptive equipment distribution is performed by staff in another health services sharing) in the previous month [5]. Such sustained such as community health or sexual health). levels of unsafe injection could be partially reduced by All clients who attended one of six NSP sites over the further increasing syringe coverage as evidence suggests study period from October 2012 to February 2013 were that approximately one in five Australian NSP clients eligible to participate. On the days that researchers were [6] and around one-third of Australian NSP pharmacy present, NSP staff informed clients about the study and scheme attendees [7] do not have adequate sterile then directed interested clients to on-site researchers injecting equipment. It is obvious that other factors, who explained to clients what the study was about and over and above the availability of sterile injecting equip- provided them with an information sheet to read. At the ment, must also contribute to the continued practices vending machine sites, a flyer was posted on the machine of sharing of needles, syringes and other injecting and the two researchers invited potential candidates to equipment among PWID [7]. participate once they had accessed their equipment. In How should NSPs increase their effectiveness in ser- addition, a small number of participants were referred to vice delivery in order to prevent transmission of hepa- the study by their peers. titis C and other BBVs in a context of high levels of equipment distribution? The risk environment frame- Ethics, consent and permissions work has generated a body of research addressing the Consenting clients completed the survey either on a physical, social, economic and policy factors that not touchscreen computer within the NSP or on paper and only enable PWID to enact safe injecting practices returned by mail. Participants were provided with $20 vou- – [8 12] but also enhance the effectiveness of NSP ser- cher to compensate them for their time. The study had eth- – vice provision [13 17]. A range of factors has been found ics approval from the Human Research Ethics Committee to impinge on the capacity of PWID to reduce a broad at the University of New South Wales (HC12128) and rele- spectrum of injection-related injuries and harms. The key vant health authorities (2011/11/4.6/(3413)). factors in the literature include socioeconomic disadvantage (poverty, unemployment, homelessness and dependence on social welfare dependence), health constraints Survey instruments and measurements due to both physical and mental illnesses [18–26] and Risky drug injection practices cognitive-behavioural factors [27, 28], as well as injected- Risky injecting practices in the past month was measured related contextual factors (e.g., injecting in public spaces) by eight recoded, binary variables: (1) reuse of any needles [29–31]) and peer networks [32–34]. or syringes (used by oneself or others); (2) receptive injec- The development and current implementation of NSP tion (injected by others rather than self-injection); (3) re- policies and service delivery models in Australia, how- ceptive sharing of any needles or syringes; (4) receptive ever, has primarily drawn upon BBV prevention princi- sharing of any injecting equipment other than needles or ples that privilege increased equipment distribution syringes; (5) distributive sharing of any needles; (6) dis- rather than addressing a range of client-prioritised needs tributive sharing of any injecting equipment other than [35–38]. In this exploratory paper, we aim to better needles or syringes; (7) injection at least once per day; and understand how NSPs could potentially enhance their (8) injection at public spaces (e.g., toilets or parks). service impact by addressing a range of client-focused needs. To achieve this, rather than adopting the more Demographic, general health and wellbeing and injection- conventional approach of soliciting independent predic- specific indicators tors of unsafe injection, we used an innovative client Apart from age, gender, country of birth (recoded into segmentation approach to investigate factors empirically Australian-born vs. other), sexual minority status (recoded differentiating NSP clients on the basis of their potential into heterosexual vs. other), Aboriginal or Torres Strait service needs including but not limited to obtaining ster- Islander (ATSI) status (yes or no) and highest level of edu- ile injecting equipment. cation attainment (recoded into up to year 10 high school only vs. other), key demographic indicators included: so- Methods cial welfare dependence (recoded into social welfare being Convenience sampling was used across six NSP sites the main source of current income vs. other); and history within a region of Sydney, Australia. The recruitment of imprisonment (yes or no). Treloar et al. Harm Reduction Journal (2016) 13:18 Page 3 of 8

Key indicators of general health and wellbeing included: particular, these 18 indicator variables were chosen as poor self-rated health (recoded into being ‘fair’ or ‘poor’ vs. they directly matched with major services that are or other); daily life stress (recoded into having more than two could be provided or facilitated by NSPs, namely, supply major life stressors in the past 12 months vs. other); history of clean needles, syringes and other injecting equipment, of any diagnoses of mental illness ever (yes or no); and ex- management of drug addiction (including treatment), istence of comorbidities (i.e., chronic physical or mental peer education/support, provision of primary health care complaints) other than hepatitis C infection (yes or no). and referral to other human services. Of note, age, gen- Injection-related specific indicators included a history der, country of birth, sexual minority, indigenous status of injection-caused injuries not related to hepatitis C and education were deliberately excluded from factor (recoded into having more than two health problems vs. analysis as these six variables were considered as less other); perceived current difficulties in managing drug modifiable demographic characteristics. The factor ana- use (yes or no); having ‘some’, ‘most’ or ‘all’ friends being lysis used a principle component extraction method, PWID (yes or no); and having spent ‘some’, ‘most’ or ‘all’ followed by a Varimax rotation and a Kaiser Normalisa- current free time with other PWID (yes or no). tion procedure. To be parsimonious, the threshold of keeping extracted variables in the final factors was set Data analysis where the minimum factor loadings were 0.40 and mini- To identify major clustering variables that potentially mum Eigenvalue was 1.5. Standardised factor scores (i.e., differentiated participants, exploratory, latent factor ana- mean = 0, sd = 1), produced by the factor analysis, were lysis was used in the first step. Based on our extensive used in the next step. For clustering purposes, each fac- knowledge from the literature and previous findings tor score was further dichotomised into >0 and ≤0(as from this project [39–41], 18 dummy coded variables weighted sums) to indicate substantial differences be- (Table 1) were selected for exploratory factor analysis. In tween participants according to each extracted factor. In

Table 1 Key factors that differentiated the sample Whole sample Factor 1a Factor2b Factor 3c (n = 236) ‘recent risky ‘disadvantage ‘drug use injection’ and disability’ milieu’ n (%) 1. >2 major life stressors in the past 12 monthsd 96 40.7 2. ‘Some’, ‘most’ or ‘all’ current friends being PWIDd 140 59.3 3. Any current, non-hepatitis C, chronic comorbidity 129 54.7 0.672 4. Ever diagnosed with a mental illness 141 59.7 0.567 5. Social welfare as a main source of current income 171 72.5 0.545 6. >2 injection-caused injuries ever 94 39.8 0.498 7. Current health being only ‘fair’ or ‘poor’ 80 33.9 0.469 8. Perceived current difficulties in managing drug use 130 55.1 0.528 9. Spending ‘some’, ‘most’ or ‘all’ current free time with other PWID 125 53.0 0.493 10. Ever imprisoned 153 64.8 0.485 In the past month 11. Distributive sharing of any other injecting equipmente 66 28.0 0.722 12. Distributive sharing of any needles 47 19.9 0.715 13. Receptive needle or syringe sharing 47 19.9 0.640 14. Receptive injection (i.e., injection by others) 47 19.9 0.585 15. Drug injection at any public spaces (e.g., toilets or parks) 64 27.1 0.525 16. Receptive sharing of any other injecting equipmente 91 38.6 0.499 17. Reuse of any (prior used) needles or syringes 73 30.9 0.475 18. Injection ≥1 per day 101 42.8 0.437 aExplained 20.2 % of total variance; Eigenvalue (rotated solution) = 3.63 bExplained 12.5 % of total variance; Eigenvalue (rotated solution) = 2.26 cExplained 8.3 % of total variance; Eigenvalue (rotated solution) = 1.50 dDropped from factor analysis eExcluding needles or syringes Treloar et al. Harm Reduction Journal (2016) 13:18 Page 4 of 8

the final step, participants were clustered according to previous month. As shown in Table 1, over half of par- different combinations of dichotomised factor scores, ticipants (n = 130, 55.1 %) perceived difficulties in man- reflecting the range of NSP client respondents with vari- aging drug use by responding ‘strongly agree’ or ‘agree’ ous degrees of injection and non-injection-related vul- to the statement ‘I find it difficult to manage my drug nerabilities. All data analysis was performed in IBM use in everyday life’. Second, of the list listed injuries SPSS Statistics 22. ever caused by drug injection, the four most commonly reported ones were (in descending order): collapsed Results veins (n = 123, 52.1 %); scarring or bruising (n =120, Demographic indicators 50.8 %); swelling of the hands or feet (n = 90, 38.1 %); and Of the 236 participants included in the analysis, 47.9 % localised infections (e.g., abscesses or cellulitis, n =87, (n = 113) was recruited from Primary NSP with co- 36.9 %). As the median of total injuries was 2 (mean = located vending machine sites, a further 30.9 % (n = 73) 2.20; sd = 1.82; min = 0; max = 8), the injection-caused from stand-alone Primary NSP sites and the rest (n = 50, injury indicator used 2 as the cut-off (i.e., median-split) 21.2 %) from secondary NSP sites. whereby approximately 40 % (n = 94) reported more than The overall participant profile has been published else- two injuries ever (Table 1). Furthermore, this group had where [41]. Briefly, close to two-thirds of the sample extensive social connections with other PWID, which was were male (n = 153) aged 39 years (sd = 9.5) and were indicated by over half of participants having ‘some,’ ‘most’ predominately Australian-born (n = 212, 89.9 %). A small or ‘all’ friends being PWID (n = 140, 59.3 %) and having proportion self-identified as being other than heterosex- spent ‘some’, ‘most’ or ‘all’ free time with other PWID ual (n = 28, 11.9 %) or being indigenous (n = 52, 22.0 %; (n = 125, 53.0 %) (Table 1). noting that this is an over-representation based on gen- Eight indicators of risky injection practices were used eral population). This group had other notable sociode- in the survey. In the previous month, as shown in mographic disadvantage. As a group, the education Table 1, the most commonly reported risky practice was attainment level was low where only 55 participants injection at least once per day (n = 101, 42.8 %); followed (23.3 %) had completed high school beyond year 10 (i.e., by receptive sharing of any injecting equipment other more than 10-year formal schooling), over 70 % of par- than needles or syringes (n = 91, 38.6 %) and reuse of ticipants (n = 171) reported government subsidies as the any needles or syringes that had been used by oneself or main source of their current income and close to two others prior (n = 73, 30.9 %). Risky injection practices thirds reported a history of imprisonment (n = 153, that were less common but still reported by a consider- 64.8 %) (Table 1). able proportion of participants included distributive sharing of any injection equipment other than needles General health and wellbeing indicators or syringes (n = 66, 28.0 %) and injection at public spaces Apart from varied degrees of sociodemographic disad- (n = 64, 27.1 %). The three least common risky practices vantage, participants’ health and wellbeing were also in the previous month were distributive sharing of nee- sub-optimal. As shown in Table 1, a third (n = 80, dles, receptive sharing of needles or syringes and recep- 33.9 %) rated their own health being generally ‘fair’ or tive injection with each practice being reported by 47 ‘poor’, over half of participants reporting ever being diag- participants (19.9 %). nosed with a mental illness (n = 141, 59.7 %) or having a current comorbidity (physical or mental) unrelated to Extracted cluster factors hepatitis C infection (n = 129, 54.7 %). In the previous The factor analysis produced three factors from 16 vari- 12 months, of the 14 listed major life stressors in the ables, accounting for 41.1 % of total variance (standard survey, the three most common ones were (in descend- Cronbach alpha = 0.72). The first factor (F1) was labelled ing order): any alcohol or drug problems (n = 125, ‘recent risky injection’ with an Eigenvalue of 3.63 after ro- 53.0 %); trouble with the police (n = 76, 32.2 %) and tation and explained 20.2 % of variance. As shown in mental conditions (n = 69, 29.2 %). As the median of Table 1, F1 consisted of all eight risky injection indicators total stressors was 2 (mean = 2.59; sd = 2.09; min = 0; and were listed in a descending order based on factor max = 10), the mental stress indicator used 2 as the loadings. The second factor (F2) was labelled ‘disadvan- cut-off (i.e., median-split) whereby approximately tage and disability’ with an Eigenvalue of 2.26 after rota- 40 % (n = 96) reported more than two stressors in the tion and explained 12.5 % of variance. Also listed in a previous 12 months (Table 1). descending order based on factor loadings, F2 consisted of four health indicators (current comorbidity, history of any Injection- specific indicators mental illness diagnoses, multiple injection-caused injuries The average age of initiation was 19.9 years (sd = 7.0) and poor health in general) and one sociodemographic in- with a majority (n = 218, 92.4 %) having injected in the dictor (social welfare dependence). The third factor (F3) Treloar et al. Harm Reduction Journal (2016) 13:18 Page 5 of 8

was labelled ‘drug use milieu’ with an Eigenvalue of 1.50 In other words, according to the three cluster factors, after rotation and explained 8.4 % of variance. Again, participants could be classified into the following eight, listed in a descending order based on factor loadings, F3 mutually exclusive groups: consisted of two injection-specific indicators (perceived difficulties in drug management and having spent a con- No substantial vulnerability (n = 40, 16.9 % in total) siderable proportion of free time with other PWID) and Single substantial vulnerability (n = 91, 38.6 % in one sociodemographic indicator (history of any imprison- total), which can be further divided into three ment). Having a considerable proportion of PWID friends groups: recent risky injection only (n = 20, 8.4 %), and experiencing multiple life stressors in the previous social disadvantage and disability only (n = 38, 12 months failed to load onto any cluster factors. 16.1 %) and drug use milieu only (n = 33, 14.0 %) Double substantial vulnerability (n = 84, 35.6 % in Client segmentation total), which can be further divided into three The three cluster factors were further dichotomised groups: both recent risky injection and social on the basis of their standardised factor scores (F1: me- disadvantage and disability (n = 19, 8.1 %); both dian = −0.34; min = −1.25; max = 2.84; F2: median = −0.34; recent risk injection and drug use milieu (n = 22, min = −1.25; max = 2.84; F3: median = 0; min = −2.73; 9.3 %); both social disadvantage and disability and max = 2.12, respectively). As shown in Fig. 1, for recent drug use milieu (n = 43, 18.2 %) risky injection, 34.7 % (n = 82) of participants were Triple substantial vulnerability (n = 21, 8.9 %) regarded as having ‘above the average’ vulnerability. For this particular sub-group with a higher vulnerability in Discussion recent risky injection, 8.4 % (n = 20) had only F1 > 0 (i.e., Using exploratory latent factor analysis, this paper shows single substantial vulnerability); 17.4 % (n = 41) had F1 > 0 that NSP clients can be essentially differentiated on as well as F2 > 0 or F3 > 0 (i.e., double substantial vulner- three distinctive domains: risky injection, socioeconomic ability); and 8.9 % (n = 21) had all three scores above 0 disadvantage and physical and/or mental comorbidities (i.e., triple substantial vulnerability). For social disadvan- and injection-related context/milieu. This key finding tage and disability, 51.3 % (n = 121) of participants were suggests that in countries like Australia where the supply regarded as having ‘above the average’ vulnerability, of sterile injecting equipment supply is already high, there which could be further divided into 16.1 % (n = 38) is an urgent need for NSPs to deliver more client-oriented with single substantial vulnerability (i.e., only F2 > 0); services that respond to clients’ real-life priorities and cir- 26.3 % (n = 62) with double substantial vulnerability cumstances, even if some of these fall outside of the cur- (i.e., F2 > 0 plus F1 > 0 or F3 > 0) and 8.9 % (n = 21) rently defined core business of NSP services. triple substantial vulnerability. For drug use milieu, In the 2014 WHO guidelines on HIV prevention, diag- 50.4 % of participants (n = 119) were regarded as hav- nosis, treatment and care for key priority groups, including ‘above the average’ vulnerability, which can be ing people who inject drugs, attention was paid to the further divided into 14.0 % (n = 33) with single sub- prevention and management of other co-infections and stantial vulnerability (i.e., only F3 > 0); 27.5 % (n = 65) comorbidities, including mental health conditions [42]. with double substantial vulnerability (i.e., F3 > 0 plus While this is a welcome advance in understanding, other F1 > 0 or F2 > 0) and 8.9 % (n = 21) with triple sub- life priorities, screening and management of comorbidities stantial vulnerability (Fig. 1). such as mental health were not positioned as associated

Fig. 1 Client segmentation: indices of substantial vulnerability. F1 recent risky injection, F2 disadvantage and disability, F3 drug use milieu Treloar et al. Harm Reduction Journal (2016) 13:18 Page 6 of 8

with risk of HIV transmission (in terms of practice) but as those classified as having substantial vulnerability in re- related to adherence to HIV treatment. This reflects the cent risky injection, a majority (62 out of 82 participants) nascent literature in relation to understanding broader were also substantially vulnerable in the socioeconomic factors that can influence risk practice among PWID. and health domain and/or in the drug use milieu domain In this paper, just over one third of NSP clients (n =82, (for example, involved in condensed PWID networks). 34.7 %) were classified as having substantial vulnerability More importantly, for clients with substantial vulnerability in recent risky injection practices, the smallest proportion across all three domains (8.9 %), NSPs could provide or fa- of all three domains. This is consistent with various NSP cilitate better client-oriented services that also support ef- surveillance data in Australia where about one quarter to forts to reduce BBV risk in the long-run. one-third PWID engage in risky injection practices during There are a number of limitations to this study. Our a specified period [5]. We argue that while increased dis- survey was conducted in one region of Sydney with tribution of sterile injection equipment is necessary, it is about 80 % of participants recruited from Primary NSP not sufficient to reduce risky injection practice. Indeed, sites. Therefore, the findings from this convenience sam- these data show that measures of vulnerability relating to ple are not generalisable to other PWID, particularly drug milieu and social disadvantage were larger than those those who do not regularly attend NSP services or other relating to injecting risk, when examined individually and sites with different models of NSP delivery. The sample in combination, suggesting that clients’ concerns and size may have limited the statistical power to explore needs are multiple and not necessarily prioritise BBV risk. other measures and detect more nuanced differences. In attempting to understand how NSP could be pro- For future research, a larger sample size and more com- vided differently and to meet client needs, we demon- prehensive measures should be considered to further il- strated that NSP clients could be classified into several luminate service needs and broader issues affecting the subgroups pertaining to various degrees of vulnerability lives of PWID. For example, measures of daily life in and across each domain. This important finding chal- stressors in the past 12 months could not capture expe- lenges the conventional assumption that NSP services riences of trauma during early childhood or sustained are best oriented towards an ‘average’ client who could through adulthood. be representative of the entire NSP client population [36]. Instead, our paper shows that it is important to ad- Conclusions dress client heterogeneity based on broader factors that For more than two decades, Australia has maintained a reflect social determinants of health [43] and indices of network of publicly funded and other privately operated risky injection practices. Our innovative client segmenta- NSP services with a focus on dispensing sterile injecting tion approach based on the exploratory factor analysis equipment, accompanied by BBV-related health promo- revealed that while over half of NSP clients in this sam- tion. Despite this, sharing of injecting equipment has ple (55.5 %) could be classified as having either no sub- persisted at concerning levels [5]. This paper supports stantial vulnerability across all three domains or only the growing body of literature suggesting that expanding one substantial vulnerability in any of the three domains, the volume of equipment distribution alone will not be the rest (i.e., 44.5 % of the sample) could be considered sufficient to eradicate risky injection [7, 25, 26]. This as having double or triple substantial vulnerability. For paper extends the literature by pointing out the import- future NSP service modelling, policy making and re- ant role that more client-tailored NSP service provision source allocation planning, our findings highlight the can play to effectively reduce BBV transmission by fur- importance of taking into account not only clients with ther addressing critical social disadvantage and improv- less demanding service needs (that is, 55 % of this sam- ing adverse health and wellbeing status. ple) but also those with chronic and more complex needs, particularly those with substantial vulnerabilities Abbreviations on aspects of life not related only to injecting drug use. BBV, blood-borne viruses; NSP, Needle and Syringe Programmes; PWID, This paper also sheds new light on the clustered rather people who inject drugs than randomly distributed nature of key indicators. Pre- Acknowledgements vious findings from this project demonstrated a signifi- We acknowledge our partners in the health authority area who funded this cant relationship between recent risky injection practices project. We thank the participants for their time and commitment to the research process. The Centre for Social Research in Health is supported by a and perceived discrimination against PWID by NSP grant from the Australian Government Department of Health and Ageing. workers [41]. We argued that this relationship could be We thank Kenneth Yates and Peter Hull for their contributions in the data mediated by mental health, that is, the synergistic rela- collection. tionship between perceived stigma/discrimination and Funding mental health is important in understanding BBV risk Funding for this project was provided by NSW Ministry of Health and the practice. In the current paper, we demonstrated that of local health districts involved. The funding body was funding involved via a Treloar et al. Harm Reduction Journal (2016) 13:18 Page 7 of 8

representative on the project’s advisory committee. This committee 15. Rich J, Strong L, Towe C, McKenzie M. Obstacles to Needle Exchange commented on proposals for the design of the study and collection, Participation in Rhode Island. J Acquir Immune Defic Syndr. 1999;21:396–400. analysis, and interpretation of data. 16. Taussig J, Weinstein B, Burris S, Jones T. Syringe laws and pharmacy regulations are structural constraints on HIV prevention in the US. AIDS. 2000;14(S1):S47–51. Availability of data and materials 17. Wood E, Kerr T, Small W, Jones J, Schechter M, Tyndall M. The impact of a The data will not be shared given the sensitive nature of the material within police presence on access to needle exchange programs. [Letter]. J Acquir it. Further, ethics approval for this project did not include provision for Immune Defic Syndr. 2003;34(1):116–8. sharing of data. 18. Craib K, Spitall P, Patel S, Christian W, Moniruzzaman A, Pearce M. Cedar Project Partnership. Prevalence and incidence of hepatitis C virus infection Authors’ contributions among Aboriginal young people who use drugs: results from the Cedar CT and LM designed the original study. All authors designed the analytical Project. Open Med. 2009;3(4):E220–7. approach. HW and LM carried out the analyses. CT drafted the manuscript. 19. Heinzerling K, Jral A, Flynn N, Anderson R, Scott A, Gilbert M, et al. Unmet All authors read and approved the final manuscript. need for recommended preventive health services among clients of California syringe exchange programs: Implications for quality improvement. Competing interests Drug Alcohol Depend. 2006;81:167–78. The authors declare that they have no competing interests. 20. Islam M, Topp L, Day C, Dawsone A, Conigrave K. The accessibility, acceptability, health impact and cost implications of primary healthcare Consent for publication outlets that target injecting drug users: A narrative synthesis of literature – Not applicable. Review. Int J Drug Policy. 2012;23(2):94 102. 21. Mandell W, Kim J, Latkin C, Suh T. Depressive symptoms, drug network, and Ethics approval and consent to participate their synergistic effect on needle-sharing behavior among street injection – The study had ethics approval from the Human Research Ethics Committee drug users. Am J Drug Alcohol Abuse. 1999;25(1):117 27. at the University of New South Wales (HC12128) and relevant health 22. McnallyS,Temple-SmithM,SievertW,PittsM.Now,laterornever?Challenges – authorities (2011/11/4.6/(3413)). associated with hepatitis C treatment. Aust N Z J Public Health. 2006;30(5):422 7. 23. Pilowsky D, Wu L-T, Burchett B, Blazer D, Ling W. Depressive Symptoms, Received: 2 December 2015 Accepted: 18 May 2016 Substance Use and HIV-Related High Risk Behaviors among Opioid- Dependent Individuals: Results from the Clinical Trials Network. Subst Use Misuse. 2011;46(14):1716–25. References 24. Riley E, Safaeian M, Strathdee S, Brooner R, Beilenson P, Vlahov D. Drug user 1. World Health Organization. Guidance on Prevention of Viral Hepatitis B and treatment referrals and entry among participants of a needle exchange – C Among People Who Inject Drugs. Geneva: World Health Organization; program. Subst Use Misuse. 2002;37:1869 86. 2012. 25. Strathdee S, Patrick D, Archibald C, Ofner M. Social determinants predict 2. World Health Organization (WHO), United Nations Office on Drugs and needle-sharing behaviour among injection drug users in Vancouver, – Crime (UNODC), & Joint United Nations Programme on HIV/AIDS (UNAIDS). Canada. Addiction. 1997;92:1339 47. Technical Guide for countries to set targets for universal access to HIV 26. Strathdee SA, Patrick DM, Archibald CP, Ofner M, Cornelisse PGA, Rekart M, prevention, treatment and care for injecting drug users. Geneva: World et al. Social determinants predict needle-sharing behaviour among injection – Health Organisation; 2009. drug users in Vancouver, Canada. Addiction. 1997;92(10):1339 47. doi:10. 3. Degenhardt L, Mathers B, Vickerman P, Rhodes T, Latkin C, Hickman M. 1111/j.1360-0443.1997.tb02852.x. Prevention of HIV infection for people who inject drugs: Why individual, 27. Wagner KD, Lankenau SE, Palinkas LA, Richardson JL, Chou CP, Unger JB. structural and combination approaches are needed. Lancet. 2010;376:285–301. The influence of the perceived consequences of refusing to share injection doi:10.1016/S0410-6736(10)60742-8. equipment among injection drug users: balancing competing risks. Addict – 4. Mathers B, Degenhardt L, Ali H, Wiessing L, Hickman M, Mattick R, et al. HIV Behav. 2011;36(8):835 42. doi:10.1016/j.addbeh.2011.03.014. prevention, treatment, and care services for people who inject drugs: a 28. Wagner KD, Unger JB, Bluthenthal RN, Andreeva VA, Pentz MA. Cognitive systematic review of global, regional, and national coverage. Lancet. 2010; behavioral theories used to explain injection risk behavior among injection 375(9719):1014–28. drug users: a review and suggestions for the integration of cognitive and – 5. Iversen J, Chow S, Maher L. Australian Needle and Syringe Program National environmental models. Health Educ Behav. 2010;37(4):504 32. doi:10.1177/ Data Report 2009–13. Sydney: Kirby Institute; 2014. 1090198109357319. 6. Iversen J, Topp L, Wand H, Maher L. Individual-level syringe coverage 29. Debeck K, Small W, Wood E, Li K, Montaner J, Kerr T. Public injecting among among Needle and Syringe Program attendees in Australia. Drug Alcohol a cohort of injecting drug users in Vancouver, Canada. J Epidemiol Depend. 2012;122(3):195–200. doi:10.1016/j.drugalcdep.2011.09.030. Community Health. 2009;63(1):81–6. doi:10.1136/jech.2007.069013. 7. Bryant, J., Paquette, D., & Wilson, H. Syringe Coverage in an Australian Setting: 30. Marshall, B. D. L., Kerr, T., Qi, J., Montaner, J. S. G., & Wood, E. (2010). Public Does a High Level of Syringe Coverage Moderate Syringe Sharing Behaviour? injecting and HIV risk behaviour among street-involved youth. Drug and AIDS and Behavior. 2012;16(5), 1156–1163. doi: DOI 10.1007/s10461-011-0010-3 Alcohol Dependence, 110(3), 254–258. doi: http://dx.doi.org/10.1016/j. 8. Burris S, Blankenship K, Donoghoe MC, Sherman SG, Vernick J, Case P, et al. drugalcdep.2010.01.022 Addressing the "risk environment" for injection drug users: The mysterious 31. Navarro, C., & Leonard, L. (2004). Prevalence and factors related to public case of the missing cop. Milbank Q. 2004;82(1):125–56. injecting in Ottawa, Canada: implications for the development of a trial 9. Moore D, Dietze P. Enabling environments and the reduction of drug- safer injecting facility. International Journal of Drug Policy, 15(4), 275–284. related harm: re-framing Australian policy and practice. Drug Alcohol Rev. doi: http://dx.doi.org/10.1016/j.drugpo.2004.03.003 2005;24:275–84. 32. De P, Cox J, Boivin J-F, Platt RW, Jolly AM. The importance of social networks in 10. Rhodes T. The ‘risk environment’: a framework for understanding and their association to drug equipment sharing among injection drug users: a reducing drug-related harm. Int J Drug Policy. 2002;13:85–97. review. Addiction. 2007;102(11):1730–9. doi:10.1111/j.1360-0443.2007.01936.x. 11. Rhodes T. Risk environments and drug harms: A social science for harm 33. Latkin C, Mandell W, Vlahov D, Oziemkowska M, Celentano D. People and reduction approach. Int J Drug Policy. 2009;20(3):193–201. places: behavioral settings and personal network characteristics as correlates of 12. Rhodes T, Simic M. Transition and the HIV risk environment. BMJ. 2005; needle sharing. J Acquir Immune Defic Syndr Hum Retrovirol. 1996;13(3):273–80. 331(7510):220–3. doi:10.1136/bmj.331.7510.220. 34. Lovell AM. Risking risk: the influence of types of capital and social networks 13. Bluthenthal R, Kral A, Lorvick J, Watters J. Impact of law enforcement on on the injection practices of drug users. Soc Sci Med. 2002;55(5):803–21. syringe exchange programs: A look at Oakland and San Francisco. Med 35. Kwon J, Anderson J, Kerr C, Thein H, Zhang L, Iversen J, et al. Estimating the cost- Anthropol. 1997;18(1):61–83. effectiveness of needle-syringe programs in Australia. AIDS. 2012;26:2201–10. 14. Lurie P, Drucker E. An opportunity lost: HIV infections associated with lack of 36. Kwon J, Iversen J, Maher L, Law M, Wilson D. The impact of needle and syringe a national needle-exchange programme in the USA. Lancet. 1997;349(9052): programs on HIV and HCV transmissions in injecting drug users in Australia: a 604–8. A. o. l. H. i. a. w. l. o. a. n. n.-e. p. i. t. U. L., 349(9052), 604–608. model-based analysis. J Acquir Immune DEfic Syndr. 2009;51:462–9. Treloar et al. Harm Reduction Journal (2016) 13:18 Page 8 of 8

37. NSW Ministry of Health. NSW Needle and Syringe Program: Guidelines. Sydney: NSW Ministry of Health; 2013. 38. Victorian Department of Human Services. National Needle and Syringe Programs Strategic Framework 2010–2014. Canberra: Commonwealth of Australia; 2010. 39. Treloar C, Hopwood M, Yates K, Mao L. ‘Doing the devil’s work’: emotional labour and stigma in expanding Needle and Syringe Programs. Drugs: educ prev policy. 2015;22(5):437–43. doi:10.3109/09687637.2015.1057553. 40. Treloar C, Rance J, Yates K, Mao L. Trust and people who inject drugs: The perspectives of clients and staff of Needle Syringe Programs. Int J Drug Policy. 2016;27:138–45. doi:10.1016/j.drugpo.2015.08.018. 41. Wilson H, Brener L, Mao L, Treloar C. Perceived discrimination and injecting risk among people who inject drugs attending Needle and Syringe Programmes in Sydney, Australia. Drug Alcohol Depend. 2014;144:274–8. doi:10.1016/j.drugalcdep.2014.08.018. 42. World Health Organisation. HIV prevention, diagnosis, treatment and care for key popuations: Consolidated guidelines. Geneva: World Health Organisation; 2014. 43. Grier S, Bryant C. Social marketing in public health. Annu Rev Public Health. 2005;26:319–39.

Submit your next manuscript to BioMed Central and we will help you at every step:

• We accept pre-submission inquiries • Our selector tool helps you to ﬁnd the most relevant journal • We provide round the clock customer support • Convenient online submission • Thorough peer review • Inclusion in PubMed and all major indexing services • Maximum visibility for your research

Submit your manuscript at www.biomedcentral.com/submit